Monday, 2015-05-04

« Sunday, 2015-05-03 Index Tuesday, 2015-05-05 »
*** cknight has joined #openstack-zaqar02:11
*** cknight has quit IRC02:25
*** achanda has quit IRC02:25
*** achanda has joined #openstack-zaqar03:26
*** achanda has quit IRC03:31
*** reed_ has joined #openstack-zaqar03:35
*** reed_ has quit IRC03:37
*** chandankumar has joined #openstack-zaqar04:38
*** chandankumar has quit IRC05:02
*** achanda has joined #openstack-zaqar05:02
*** chandankumar has joined #openstack-zaqar05:55
openstackgerritOpenStack Proposal Bot proposed openstack/zaqar: Imported Translations from Transifex  https://review.openstack.org/17963006:02
flaper87flwang: ping (re review)06:34
flaper87vkmc: ping06:35
flaper87you should be in CEST by now!06:35
flaper87wake UP!06:35
flaper87flwang: https://review.openstack.org/#/c/178093/ pls06:36
flaper87flwang: and https://review.openstack.org/#/c/178099/206:36
flaper87flwang: vkmc we really need those patches to move the client forward.06:37
flaper87well, I gotta fix the pooling one, will do that in a bit06:37
*** achanda has quit IRC07:23
*** achanda has joined #openstack-zaqar07:26
*** achanda has quit IRC07:37
*** dynarro has joined #openstack-zaqar07:50
*** dynarro has quit IRC07:52
*** dynarro has joined #openstack-zaqar08:04
*** dynarro has quit IRC08:25
*** _dguerri is now known as dguerri09:17
*** dguerri is now known as _dguerri09:28
openstackgerritFei Long Wang proposed openstack/zaqar-specs: Add the specs dir for Liberty  https://review.openstack.org/17967009:29
*** flwang1 has joined #openstack-zaqar09:30
openstackgerritFei Long Wang proposed openstack/zaqar-specs: Policy support  https://review.openstack.org/17967309:34
*** _dguerri is now known as dguerri09:42
*** dguerri is now known as _dguerri09:50
flwang1flaper87: around?09:53
*** _dguerri is now known as dguerri10:27
*** exploreshaifali has joined #openstack-zaqar10:40
flaper87flwang1: yes10:42
flaper87well, a bit late but  yes10:42
flwang1flaper87: https://review.openstack.org/#/c/17967010:42
flwang1liberty is open :)10:43
flwang1flaper87:  and this https://review.openstack.org/#/c/179673/ for the policy support10:43
*** dguerri is now known as _dguerri10:43
flwang1that's one of my to-do list in L10:43
flaper87flwang1: just if you review the patches I mentioned before :)10:43
flaper87:P10:43
flaper87flwang1: https://review.openstack.org/#/c/178093/10:43
flwang1flaper87: i did10:43
flaper87you -1'd it but I replied10:43
flaper87flwang1: and https://review.openstack.org/#/c/178099/210:44
flaper87pls, we really need to move those forward10:44
openstackgerritFei Long Wang proposed openstack/zaqar-specs: Policy support  https://review.openstack.org/17967310:45
flwang1flaper87: at line 198, 200, 213, they are still using zaqar-server, should they be renamed as well?10:46
flwang1https://review.openstack.org/#/c/178093/2/devstack/plugin.sh10:46
flaper87flwang1: no, that's the actuall process that is being launched10:47
flaper87I'm just renaming the service name in *devstack*10:47
flaper87not in Zaqar10:47
flwang1ok, thanks for the clarification10:48
flaper87I'll make this better in a follow up patch but for now, we should just keep it like that, IMHO10:48
flwang1done, +2ed10:49
flaper87flwang1: danke10:49
flwang1flaper87: btw, as for the v2 support in client, does it need a spec?10:51
flaper87flwang1: yes, it does.10:56
flwang1ok10:56
flaper87flwang1: lets hold off a bit until vkmc is back. I believe this is something Outreachy interns will work on10:56
flaper87I would hate taking over their work10:57
flwang1it would be nice if there is any intern can take that10:57
flwang1flaper87: FWIW, i would like to implement the policy support asap10:57
flaper87flwang1: we have 3 interns for Zaqar this round10:58
flwang1awesome10:59
*** exploreshaifali has quit IRC10:59
flaper87flwang1: ok, I'll be on spec reviews soon enough10:59
flaper87I need to sort out the gate first10:59
flwang1ok10:59
flwang1pls let me know if there is anything i can help10:59
flaper87flwang1: review all that chain of patches11:03
flwang1i'm on that11:04
flaper87flwang1: also, the last one in the chain is failing, I checked it last week and I believe I know what's failing11:04
flaper87however, it'd be good to get your input there11:04
flwang1cool11:04
flwang1I can do that on my day11:04
*** dynarro has joined #openstack-zaqar11:15
*** exploreshaifali has joined #openstack-zaqar11:56
*** exploreshaifali has quit IRC12:00
*** exploreshaifali has joined #openstack-zaqar12:00
*** exploreshaifali has quit IRC12:03
*** exploreshaifali has joined #openstack-zaqar12:03
*** exploreshaifali has quit IRC12:07
*** exploreshaifali has joined #openstack-zaqar12:08
*** exploreshaifali has quit IRC12:08
*** exploreshaifali has joined #openstack-zaqar12:15
*** exploreshaifali has quit IRC12:17
*** exploreshaifali has joined #openstack-zaqar12:18
*** shakamunyi has quit IRC12:18
*** exploreshaifali has quit IRC12:19
*** exploreshaifali has joined #openstack-zaqar12:20
*** exploreshaifali has quit IRC12:21
flaper87vkmc: yo yo yo!12:27
*** _dguerri is now known as dguerri12:38
*** dguerri is now known as _dguerri12:45
*** exploreshaifali has joined #openstack-zaqar12:48
*** exploreshaifali has quit IRC12:51
*** exploreshaifali has joined #openstack-zaqar12:52
*** exploreshaifali has quit IRC12:57
*** exploreshaifali has joined #openstack-zaqar12:57
*** exploreshaifali has quit IRC12:59
*** exploreshaifali has joined #openstack-zaqar13:00
vkmchey o/13:00
*** exploreshaifali has quit IRC13:00
flaper87vkmc: hey hey :)13:01
vkmcyes, we are going to have dynarro working on the client side13:01
vkmcadding support for all the features we have en v1_113:01
flaper87awesome!13:01
flaper87vkmc: https://review.openstack.org/#/c/178100/ <- review the ones this depends on13:02
flaper87I'm fixing it now13:02
vkmcok13:02
*** sriram has joined #openstack-zaqar13:10
vkmcflaper87, I reviewed them all13:10
thervevkmc, https://review.openstack.org/#/c/177792/ FWIW13:10
vkmctherve, thanks, I'll take a look on that13:11
therveHopefully gate will pass once the devstack changes are in13:13
flaper87therve: fixing that right now, we'll know the answer to that soon13:20
flaper87(Last famous words)13:20
*** ametts has joined #openstack-zaqar13:25
*** amitgandhinz has joined #openstack-zaqar13:26
*** chandankumar has quit IRC13:35
*** cknight has joined #openstack-zaqar13:42
flaper87vkmc: https://review.openstack.org/#/c/178093/13:45
flaper87:P13:45
vkmcflaper87, done13:55
*** malini has joined #openstack-zaqar14:00
*** _dguerri is now known as dguerri14:12
*** dguerri is now known as _dguerri14:13
*** _dguerri is now known as dguerri14:13
*** exploreshaifali has joined #openstack-zaqar14:19
therveflaper87, You should reference a bug when the gate fails14:37
flaper87therve: I know, in this case pypi failed and I don't have a bug.14:37
therveflaper87, https://bugs.launchpad.net/openstack-gate/+bug/144913614:37
openstackLaunchpad bug 1449136 in OpenStack-Gate "OpenStack pypi mirrors disconnecting connections" [Undecided,New]14:37
flaper87if you re-approve the patch, it should just re-do the check14:37
flaper87unless you're using reverify14:37
* flaper87 went lazy14:37
therve:)14:38
vkmcflaper87 is always lazy14:40
*** malini has quit IRC14:41
flaper87fun thing, the pooling patch fails because we need to create pools but zaqarclient doesn't have support for it yet. Manual pool creation it is14:43
flaper87at least until we unblock this thing14:43
flaper87I really think we should default to whatever is configured in the config file if no pools exist14:44
flaper87I thought we were already doing this14:44
vkmcnope, if pools is enabled and there are no pools14:44
vkmceverything will fail14:44
flaper87fuck that14:44
vkmc¯\_(ツ)_/¯14:45
sriramflaper87: congrats on TC! :)14:47
flaper87sriram: thanks :)14:47
vkmcflaper87, you have to tell me more about what you are working on with proton14:47
therveflaper87, It feels that skipping theses tests for now would be a good way forward14:48
*** chandankumar has joined #openstack-zaqar14:50
*** kgriffs|afk is now known as kgriffs15:00
*** mpanetta has joined #openstack-zaqar15:01
flaper87therve: I think I have a fair fix on the server side, I'll push it in  a bit15:01
* flaper87 has been jumping between calls the whole day15:01
flaper87tox tox tox15:07
flaper87that sounds like coughing15:08
vkmcflaper87, what fix do you have in mind?15:08
flaper87also, I'm still mad at Jiriu for breaking Zaqar with the latest pymongo release15:09
flaper87>.>15:09
flaper87yes yes, it's a major but arrghh15:09
flaper87:(15:09
flaper87now we need to create a mongodb six15:09
flaper87:P15:09
flaper87that said, I also have an almost-ready patch for that15:10
vkmck15:10
vkmcconsidering that we need to do a lot of work client side, skipping these tests is not a bad approach IMO15:10
*** malini has joined #openstack-zaqar15:11
flaper87the fix is the one we've been discussing for ages. If there are no pools and there's a `message_store` configured in the config file, use that as the default pool15:12
flaper87no default creation, just use what's in the config file15:12
vkmccool15:12
*** malini has quit IRC15:25
exploreshaifaliflaper87, congratulations on TC!!! :D15:32
exploreshaifalivkmc, \o/15:33
vkmc:)15:36
vkmchow are you exploreshaifali?15:36
exploreshaifalivkmc, super goooooooood :)15:36
exploreshaifalivkmc, and you?15:36
* flaper87 back15:36
exploreshaifalilooks veeeeeeeeeeeeeeeeeeeeeery busy15:36
flaper87exploreshaifali: hey there!15:36
vkmcexploreshaifali, a bit tired, but doing good as well15:36
flaper87:)15:36
exploreshaifaliflaper87, \o/15:37
exploreshaifaliflaper87, congrats!!!15:37
flaper87exploreshaifali: ah thanks :)15:37
exploreshaifaliflaper87, we will do hard pary in conf!15:38
exploreshaifaliflaper87, some work/preparation you wanna me to do for conf as part of Zaqar team?15:39
exploreshaifalilike we will have some sessions for Zaqar, right? so for that?15:39
vkmcyeah, we have to plan those15:40
exploreshaifaliit is very close, if that is not planned yet, we need to do soon15:41
exploreshaifalivkmc, please feel free to assign me any task for it15:42
vkmcsure, thanks exploreshaifali15:42
exploreshaifalilook who is thanking!!!!15:42
flaper87exploreshaifali: vkmc yeah, I'll be filling those up later today/tomorrow15:42
exploreshaifaliok ok :)15:43
vkmcflaper87, let me know so we can set up an etherpad and draft it together15:43
*** malini has joined #openstack-zaqar15:45
openstackgerritMerged openstack/zaqar: Rename service to zaqar  https://review.openstack.org/17809315:50
*** kgriffs is now known as kgriffs|afk15:59
*** dguerri is now known as _dguerri16:06
*** _dguerri is now known as dguerri16:21
*** dynarro has quit IRC16:36
*** chandankumar has quit IRC16:38
openstackgerritFlavio Percoco proposed openstack/zaqar: Enable pooling in devstack  https://review.openstack.org/17810016:55
openstackgerritMerged openstack/zaqar: Remove support for sqlalchemy in devstack  https://review.openstack.org/17809817:01
openstackgerritMerged openstack/zaqar: Update dataplane section name  https://review.openstack.org/17809917:01
flaper87w000h00017:01
flaper87now, lets see if the other one passes tests17:02
*** dguerri is now known as _dguerri17:06
*** achanda has joined #openstack-zaqar17:11
*** chandankumar has joined #openstack-zaqar17:14
therveflaper87, There is a pdb in your code17:28
therveThe gate may not like it very much17:29
flaper87fuck17:29
flaper87T_T17:29
flaper87I forgot to remove it after my last test17:29
flaper87I believe that's one of the most shameful things that can happen to a python dev17:29
flaper87and probably one of the most common17:29
therveMeh17:29
* flaper87 should run more checks before submitting17:30
therveShame is overrated17:30
openstackgerritFlavio Percoco proposed openstack/zaqar: Enable pooling in devstack  https://review.openstack.org/17810017:30
flaper87therve: ah well, yeah. I "suffer" of lack of it.... ask anyone in this channel :P17:30
therve:D17:31
openstackgerritMerged openstack/zaqar-specs: Add the specs dir for Liberty  https://review.openstack.org/17967018:01
*** exploreshaifali has quit IRC18:03
*** exploreshaifali has joined #openstack-zaqar18:03
*** exploreshaifali_ has joined #openstack-zaqar18:03
*** exploreshaifali has quit IRC18:04
*** exploreshaifali_ has quit IRC18:04
*** exploreshaifali has joined #openstack-zaqar18:04
*** exploreshaifali has quit IRC18:13
*** boris-42 has joined #openstack-zaqar18:37
openstackgerritOpenStack Proposal Bot proposed openstack/python-zaqarclient: Updated from global requirements  https://review.openstack.org/13599118:51
openstackgerritOpenStack Proposal Bot proposed openstack/zaqar: Updated from global requirements  https://review.openstack.org/17685518:52
*** kgriffs|afk is now known as kgriffs18:53
*** ametts has quit IRC19:06
*** malini has quit IRC19:29
*** flwang1 has quit IRC19:33
*** chandankumar has quit IRC19:51
openstackgerritOpenStack Proposal Bot proposed openstack/python-zaqarclient: Updated from global requirements  https://review.openstack.org/13599120:09
openstackgerritOpenStack Proposal Bot proposed openstack/zaqar: Updated from global requirements  https://review.openstack.org/17685520:10
*** VeggieMeat has quit IRC20:24
*** VeggieMeat has joined #openstack-zaqar20:24
*** exploreshaifali has joined #openstack-zaqar20:29
*** boris-42 has quit IRC20:30
*** VeggieMeat has quit IRC20:31
*** VeggieMeat_ has joined #openstack-zaqar20:31
*** boris-42 has joined #openstack-zaqar20:36
*** amitgandhinz has quit IRC20:48
*** exploreshaifali has quit IRC20:51
*** VeggieMeat_ is now known as VeggieMeat20:55
*** flwang1 has joined #openstack-zaqar21:10
*** amitgandhinz has joined #openstack-zaqar21:15
*** sriram has quit IRC21:41
flwangflaper87: ping21:54
flwangvkmc: ping21:54
vkmcflwang, pong21:54
*** _dguerri has quit IRC21:54
flwangas for using AMQP as backend, i think it's in pending, right?21:54
flwangvkmc: since we(in my team) are discussing the adopting of Trove21:56
flwangand you know based on current status, we need a separate MQ for Trove, so how to do that in production, especially in a public cloud21:56
flwangI'm asking since we're trying to figure out if it's possible to leverage Zaqar as the MQ21:57
flwangand do you think it's worth a blueprint?21:57
flwangwe don't mind if it's rejected by  Trove team, because at least we can know why it doesn't work21:58
flwangthoughts?21:58
flwangvkmc: if you think it's good to talk in Trove channel, i'm ok to pop :)21:59
vkmcso..  hmm, I'm not sure if Zaqar adjusts to Trove needs22:00
*** dguerri has joined #openstack-zaqar22:00
vkmcthe idea seems appealing to me, and certainly it could work22:00
flwangvkmc: ok, so let's split this into 2 topics22:01
vkmcbut I should try to see the edge cases in which Trove may need something more like Rabbit instead of Zaqar22:01
vkmcsure22:01
flwang1. how to deploy Trove to avoid any security issue22:02
flwangJust sent a reply to this on the dev-list, but the gist of this is that  in pretty much all of the real-world Trove deployments use their own  rabbitmq server deployed as an in cloud workload, so there is no  requirement to have communication between the guest and the  infrastructure rabbitmq server.22:02
flwangI got this from Nikhil22:02
vkmcyeah, that makes sense22:03
flwangjust like you mentioned, it's recommended to deploy a separate mq for trove, right?22:03
vkmcso, they solved the security issue by having two instances of rabbitmq22:03
vkmcyeah22:03
vkmcwe would like to have a better solution for this though22:03
flwangbut Rabbitmq doesn't support multi-tenancy22:04
vkmcsomeone deploying Trove wrongly would hit with that security issue and we cannot afford that22:04
flwangso does that mean all the instances deployed by Trove have to share the same account of RabbitMQ?22:04
vkmcI don't think multitenancy is needed in this case because we use different RabbitMQ queues for that22:04
vkmceverytime a instance is launched a new queue is created to communicate it with Trove server22:05
vkmcand yes, ATM all the instances deployed by Trove share the same account22:05
vkmcbut have different queues according to the instance22:06
flwangbut the in the agent, it will still contain the password of the mq, right?22:06
vkmcthe login credentials, no22:06
vkmcbut the guest is able to ping the rabbitmq server22:06
vkmcand use it if they want22:07
vkmcthere is a problem if we use "generic" users22:07
vkmcall across OpenStack we are using the same RabbitMQ user for everything22:07
vkmcI'm heading off for today, but I'll be tomorrow morning :)22:10
vkmco/22:10
flwangok, thanks vkmc22:10
flwanghave a good night22:10
*** cknight has quit IRC22:15
*** amitgandhinz has quit IRC22:20
*** mpanetta has quit IRC22:50
« Sunday, 2015-05-03 Index Tuesday, 2015-05-05 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!